Computational Technology Resources - CCC

D. Gunseren¹, A.A. Atik¹, N. Muhtaroglu¹, I. Ari¹ and Ö. Ertunç²

¹Computer Engineering Department, Ozyegin University, Istanbul, Turkey
²Mechanical Engineering Department, Ozyegin University, Istanbul, Turkey

doi:10.4203/ccc.2.7.2

D. Gunseren, A.A. Atik, N. Muhtaroglu, I. Ari, Ö. Ertunç, "Predicting Cases of Isotropy in Turbulence Modeling using Physics Informed Machine Learning", in B.H.V. Topping, P. Iványi, (Editors), "Proceedings of the Eleventh International Conference on Engineering Computational Technology", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 2, Paper 7.2, 2022, doi:10.4203/ccc.2.7.2

Keywords: physics informed machine learning, turbulence, anisotropy, neural networks, OpenFoam.

Abstract

Turbulence is a well-ploughed area in computational fluid dynamics (CFD). However, modern DNS-LES-RANS techniques are still computationally heavy and/or inaccurate at high Reynolds numbers. Due to lack of fine-granularity and optimality with manual tuning of model parameters, an opportunity for machine learning emerges. This paper delivers accurate turbulence models dynamically, by combining decades old scientific turbulence foundations with novel Physics Informed Machine Learning (PIML) techniques. As a starting point, we train different regression and neural network algorithms over Isotropy Cases, yet we plan to extend our work with all anisotropy cases that can be represented within the universal Lumley triangle.

download the full-text of this paper (PDF, 7 pages, 376 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 2 PROCEEDINGS OF THE ELEVENTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY Edited by: B.H.V. Topping and P. Iványi Paper 7.2 Predicting Cases of Isotropy in Turbulence Modeling using Physics Informed Machine Learning D. Gunseren¹, A.A. Atik¹, N. Muhtaroglu¹, I. Ari¹ and Ö. Ertunç² ¹Computer Engineering Department, Ozyegin University, Istanbul, Turkey ²Mechanical Engineering Department, Ozyegin University, Istanbul, Turkey doi:10.4203/ccc.2.7.2 Full Bibliographic Reference for this paper D. Gunseren, A.A. Atik, N. Muhtaroglu, I. Ari, Ö. Ertunç, "Predicting Cases of Isotropy in Turbulence Modeling using Physics Informed Machine Learning", in B.H.V. Topping, P. Iványi, (Editors), "Proceedings of the Eleventh International Conference on Engineering Computational Technology", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 2, Paper 7.2, 2022, doi:10.4203/ccc.2.7.2 Keywords: physics informed machine learning, turbulence, anisotropy, neural networks, OpenFoam. Abstract Turbulence is a well-ploughed area in computational fluid dynamics (CFD). However, modern DNS-LES-RANS techniques are still computationally heavy and/or inaccurate at high Reynolds numbers. Due to lack of fine-granularity and optimality with manual tuning of model parameters, an opportunity for machine learning emerges. This paper delivers accurate turbulence models dynamically, by combining decades old scientific turbulence foundations with novel Physics Informed Machine Learning (PIML) techniques. As a starting point, we train different regression and neural network algorithms over Isotropy Cases, yet we plan to extend our work with all anisotropy cases that can be represented within the universal Lumley triangle. download the full-text of this paper (PDF, 7 pages, 376 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions